| { | |
| "best_metric": 2.9037797451019287, | |
| "best_model_checkpoint": "/home/datta0/models/lora_final/gemma-2-9b_pct_ortho_r32/checkpoint-8", | |
| "epoch": 0.9889738430583501, | |
| "eval_steps": 8, | |
| "global_step": 384, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.002575452716297787, | |
| "grad_norm": 3.382777214050293, | |
| "learning_rate": 1.25e-05, | |
| "loss": 2.1767, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010301810865191148, | |
| "grad_norm": 3.251157760620117, | |
| "learning_rate": 5e-05, | |
| "loss": 2.2653, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.020603621730382295, | |
| "grad_norm": 3.237678050994873, | |
| "learning_rate": 0.0001, | |
| "loss": 2.3352, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.020603621730382295, | |
| "eval_loss": 2.9037797451019287, | |
| "eval_runtime": 650.9667, | |
| "eval_samples_per_second": 0.376, | |
| "eval_steps_per_second": 0.376, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.03090543259557344, | |
| "grad_norm": 5.404860019683838, | |
| "learning_rate": 9.997266286704631e-05, | |
| "loss": 5.7565, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.04120724346076459, | |
| "grad_norm": 0.5222948789596558, | |
| "learning_rate": 9.989068136093873e-05, | |
| "loss": 11.3417, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.04120724346076459, | |
| "eval_loss": 11.908318519592285, | |
| "eval_runtime": 615.5742, | |
| "eval_samples_per_second": 0.398, | |
| "eval_steps_per_second": 0.398, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.05150905432595573, | |
| "grad_norm": 0.5839648842811584, | |
| "learning_rate": 9.975414512725057e-05, | |
| "loss": 11.9397, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06181086519114688, | |
| "grad_norm": 0.2988390028476715, | |
| "learning_rate": 9.956320346634876e-05, | |
| "loss": 11.9918, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.06181086519114688, | |
| "eval_loss": 11.977351188659668, | |
| "eval_runtime": 124.2673, | |
| "eval_samples_per_second": 1.972, | |
| "eval_steps_per_second": 1.972, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.07211267605633803, | |
| "grad_norm": 0.4247038662433624, | |
| "learning_rate": 9.931806517013612e-05, | |
| "loss": 11.9842, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.08241448692152918, | |
| "grad_norm": 0.1009724959731102, | |
| "learning_rate": 9.901899829374047e-05, | |
| "loss": 11.9549, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.08241448692152918, | |
| "eval_loss": 11.967530250549316, | |
| "eval_runtime": 155.0418, | |
| "eval_samples_per_second": 1.58, | |
| "eval_steps_per_second": 1.58, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.09271629778672032, | |
| "grad_norm": 0.15956225991249084, | |
| "learning_rate": 9.86663298624003e-05, | |
| "loss": 11.9835, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.10301810865191147, | |
| "grad_norm": 0.08006568253040314, | |
| "learning_rate": 9.826044551386744e-05, | |
| "loss": 11.974, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.10301810865191147, | |
| "eval_loss": 11.973628997802734, | |
| "eval_runtime": 119.2748, | |
| "eval_samples_per_second": 2.054, | |
| "eval_steps_per_second": 2.054, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11331991951710262, | |
| "grad_norm": 0.0759950578212738, | |
| "learning_rate": 9.780178907671789e-05, | |
| "loss": 11.971, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.12362173038229377, | |
| "grad_norm": 0.06046704575419426, | |
| "learning_rate": 9.729086208503174e-05, | |
| "loss": 11.9403, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.12362173038229377, | |
| "eval_loss": 11.946824073791504, | |
| "eval_runtime": 176.6154, | |
| "eval_samples_per_second": 1.387, | |
| "eval_steps_per_second": 1.387, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.1339235412474849, | |
| "grad_norm": 0.07782606780529022, | |
| "learning_rate": 9.672822322997305e-05, | |
| "loss": 11.9184, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.14422535211267606, | |
| "grad_norm": 0.04233577847480774, | |
| "learning_rate": 9.611448774886924e-05, | |
| "loss": 11.9321, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.14422535211267606, | |
| "eval_loss": 11.880873680114746, | |
| "eval_runtime": 115.3733, | |
| "eval_samples_per_second": 2.124, | |
| "eval_steps_per_second": 2.124, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.1545271629778672, | |
| "grad_norm": 0.042002614587545395, | |
| "learning_rate": 9.545032675245813e-05, | |
| "loss": 11.8934, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.16482897384305836, | |
| "grad_norm": 0.047231078147888184, | |
| "learning_rate": 9.473646649103818e-05, | |
| "loss": 11.876, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.16482897384305836, | |
| "eval_loss": 11.821806907653809, | |
| "eval_runtime": 122.0477, | |
| "eval_samples_per_second": 2.007, | |
| "eval_steps_per_second": 2.007, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.1751307847082495, | |
| "grad_norm": 0.05491908639669418, | |
| "learning_rate": 9.397368756032445e-05, | |
| "loss": 11.789, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.18543259557344063, | |
| "grad_norm": 0.05340013653039932, | |
| "learning_rate": 9.316282404787871e-05, | |
| "loss": 11.7886, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.18543259557344063, | |
| "eval_loss": 11.734496116638184, | |
| "eval_runtime": 177.6856, | |
| "eval_samples_per_second": 1.379, | |
| "eval_steps_per_second": 1.379, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.1957344064386318, | |
| "grad_norm": 0.04357607662677765, | |
| "learning_rate": 9.230476262104677e-05, | |
| "loss": 11.7139, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.20603621730382293, | |
| "grad_norm": 0.05346447974443436, | |
| "learning_rate": 9.140044155740101e-05, | |
| "loss": 11.6471, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.20603621730382293, | |
| "eval_loss": 11.623627662658691, | |
| "eval_runtime": 116.4544, | |
| "eval_samples_per_second": 2.104, | |
| "eval_steps_per_second": 2.104, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2163380281690141, | |
| "grad_norm": 0.071262426674366, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 11.6208, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.22663983903420523, | |
| "grad_norm": 0.04146931692957878, | |
| "learning_rate": 8.945702546981969e-05, | |
| "loss": 11.5982, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.22663983903420523, | |
| "eval_loss": 11.371801376342773, | |
| "eval_runtime": 163.8682, | |
| "eval_samples_per_second": 1.495, | |
| "eval_steps_per_second": 1.495, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.23694164989939637, | |
| "grad_norm": 0.04036790132522583, | |
| "learning_rate": 8.842005554284296e-05, | |
| "loss": 11.6333, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.24724346076458753, | |
| "grad_norm": 0.03238769620656967, | |
| "learning_rate": 8.73410738492077e-05, | |
| "loss": 11.7088, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.24724346076458753, | |
| "eval_loss": 11.679157257080078, | |
| "eval_runtime": 128.5572, | |
| "eval_samples_per_second": 1.906, | |
| "eval_steps_per_second": 1.906, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.2575452716297787, | |
| "grad_norm": 0.06817040592432022, | |
| "learning_rate": 8.622126023955446e-05, | |
| "loss": 11.7336, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2678470824949698, | |
| "grad_norm": 0.042927809059619904, | |
| "learning_rate": 8.506183921362443e-05, | |
| "loss": 11.7296, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.2678470824949698, | |
| "eval_loss": 11.688301086425781, | |
| "eval_runtime": 164.6581, | |
| "eval_samples_per_second": 1.488, | |
| "eval_steps_per_second": 1.488, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.27814889336016096, | |
| "grad_norm": 0.057764846831560135, | |
| "learning_rate": 8.386407858128706e-05, | |
| "loss": 11.681, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.28845070422535213, | |
| "grad_norm": 0.04680996388196945, | |
| "learning_rate": 8.262928807620843e-05, | |
| "loss": 11.6508, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.28845070422535213, | |
| "eval_loss": 11.441954612731934, | |
| "eval_runtime": 400.4322, | |
| "eval_samples_per_second": 0.612, | |
| "eval_steps_per_second": 0.612, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.29875251509054324, | |
| "grad_norm": 0.0837872251868248, | |
| "learning_rate": 8.135881792367686e-05, | |
| "loss": 11.4868, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 0.3090543259557344, | |
| "grad_norm": 1.4256629943847656, | |
| "learning_rate": 8.005405736415126e-05, | |
| "loss": 10.7655, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.3090543259557344, | |
| "eval_loss": 8.817360877990723, | |
| "eval_runtime": 405.6339, | |
| "eval_samples_per_second": 0.604, | |
| "eval_steps_per_second": 0.604, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.31935613682092556, | |
| "grad_norm": 0.6819819211959839, | |
| "learning_rate": 7.871643313414718e-05, | |
| "loss": 9.2146, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 0.3296579476861167, | |
| "grad_norm": 1.2133327722549438, | |
| "learning_rate": 7.734740790612136e-05, | |
| "loss": 8.5075, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.3296579476861167, | |
| "eval_loss": 9.056794166564941, | |
| "eval_runtime": 582.8769, | |
| "eval_samples_per_second": 0.42, | |
| "eval_steps_per_second": 0.42, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.33995975855130783, | |
| "grad_norm": 1.0759503841400146, | |
| "learning_rate": 7.594847868906076e-05, | |
| "loss": 8.4476, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.350261569416499, | |
| "grad_norm": 0.6876762509346008, | |
| "learning_rate": 7.452117519152542e-05, | |
| "loss": 8.912, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.350261569416499, | |
| "eval_loss": 9.416214942932129, | |
| "eval_runtime": 404.0983, | |
| "eval_samples_per_second": 0.606, | |
| "eval_steps_per_second": 0.606, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.36056338028169016, | |
| "grad_norm": 0.4839273691177368, | |
| "learning_rate": 7.30670581489344e-05, | |
| "loss": 10.0255, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.37086519114688127, | |
| "grad_norm": 0.5928329229354858, | |
| "learning_rate": 7.158771761692464e-05, | |
| "loss": 11.0052, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.37086519114688127, | |
| "eval_loss": 10.347317695617676, | |
| "eval_runtime": 589.7264, | |
| "eval_samples_per_second": 0.415, | |
| "eval_steps_per_second": 0.415, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.38116700201207243, | |
| "grad_norm": 0.6406193971633911, | |
| "learning_rate": 7.008477123264848e-05, | |
| "loss": 10.2362, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 0.3914688128772636, | |
| "grad_norm": 0.6814000606536865, | |
| "learning_rate": 6.855986244591104e-05, | |
| "loss": 9.103, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.3914688128772636, | |
| "eval_loss": 9.645101547241211, | |
| "eval_runtime": 412.3792, | |
| "eval_samples_per_second": 0.594, | |
| "eval_steps_per_second": 0.594, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.4017706237424547, | |
| "grad_norm": 1.015820860862732, | |
| "learning_rate": 6.701465872208216e-05, | |
| "loss": 9.8991, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.41207243460764587, | |
| "grad_norm": 1.2941161394119263, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 8.9631, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.41207243460764587, | |
| "eval_loss": 8.64919662475586, | |
| "eval_runtime": 589.6714, | |
| "eval_samples_per_second": 0.415, | |
| "eval_steps_per_second": 0.415, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.42237424547283703, | |
| "grad_norm": 0.9186323881149292, | |
| "learning_rate": 6.387014543809223e-05, | |
| "loss": 9.4897, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 0.4326760563380282, | |
| "grad_norm": 1.1817784309387207, | |
| "learning_rate": 6.227427435703997e-05, | |
| "loss": 9.9634, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.4326760563380282, | |
| "eval_loss": 9.44008731842041, | |
| "eval_runtime": 410.3179, | |
| "eval_samples_per_second": 0.597, | |
| "eval_steps_per_second": 0.597, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.4429778672032193, | |
| "grad_norm": 0.6623127460479736, | |
| "learning_rate": 6.066498153718735e-05, | |
| "loss": 9.3577, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 0.45327967806841046, | |
| "grad_norm": 0.612003743648529, | |
| "learning_rate": 5.90440267166055e-05, | |
| "loss": 9.814, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.45327967806841046, | |
| "eval_loss": 10.474808692932129, | |
| "eval_runtime": 407.6119, | |
| "eval_samples_per_second": 0.601, | |
| "eval_steps_per_second": 0.601, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.4635814889336016, | |
| "grad_norm": 1.1024558544158936, | |
| "learning_rate": 5.74131823855921e-05, | |
| "loss": 10.5504, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.47388329979879273, | |
| "grad_norm": 0.6612622737884521, | |
| "learning_rate": 5.577423184847932e-05, | |
| "loss": 10.507, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.47388329979879273, | |
| "eval_loss": 10.191047668457031, | |
| "eval_runtime": 619.1618, | |
| "eval_samples_per_second": 0.396, | |
| "eval_steps_per_second": 0.396, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.4841851106639839, | |
| "grad_norm": 0.6979438662528992, | |
| "learning_rate": 5.4128967273616625e-05, | |
| "loss": 10.0648, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 0.49448692152917506, | |
| "grad_norm": 1.1083264350891113, | |
| "learning_rate": 5.247918773366112e-05, | |
| "loss": 9.6613, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.49448692152917506, | |
| "eval_loss": 9.220105171203613, | |
| "eval_runtime": 407.7177, | |
| "eval_samples_per_second": 0.601, | |
| "eval_steps_per_second": 0.601, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.5047887323943662, | |
| "grad_norm": 1.5402456521987915, | |
| "learning_rate": 5.0826697238317935e-05, | |
| "loss": 9.0613, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 0.5150905432595574, | |
| "grad_norm": 0.9043979644775391, | |
| "learning_rate": 4.917330276168208e-05, | |
| "loss": 9.0448, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5150905432595574, | |
| "eval_loss": 10.391300201416016, | |
| "eval_runtime": 612.3848, | |
| "eval_samples_per_second": 0.4, | |
| "eval_steps_per_second": 0.4, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5253923541247485, | |
| "grad_norm": 1.0502400398254395, | |
| "learning_rate": 4.7520812266338885e-05, | |
| "loss": 10.3086, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 0.5356941649899396, | |
| "grad_norm": 1.2477459907531738, | |
| "learning_rate": 4.5871032726383386e-05, | |
| "loss": 9.4984, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.5356941649899396, | |
| "eval_loss": 8.543373107910156, | |
| "eval_runtime": 401.1613, | |
| "eval_samples_per_second": 0.611, | |
| "eval_steps_per_second": 0.611, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.5459959758551308, | |
| "grad_norm": 1.947261929512024, | |
| "learning_rate": 4.4225768151520694e-05, | |
| "loss": 7.6535, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 0.5562977867203219, | |
| "grad_norm": 1.2403180599212646, | |
| "learning_rate": 4.2586817614407895e-05, | |
| "loss": 7.4393, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.5562977867203219, | |
| "eval_loss": 8.435011863708496, | |
| "eval_runtime": 596.6953, | |
| "eval_samples_per_second": 0.411, | |
| "eval_steps_per_second": 0.411, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.566599597585513, | |
| "grad_norm": 0.40848633646965027, | |
| "learning_rate": 4.095597328339452e-05, | |
| "loss": 9.1596, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.5769014084507043, | |
| "grad_norm": 0.33752307295799255, | |
| "learning_rate": 3.933501846281267e-05, | |
| "loss": 10.0883, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.5769014084507043, | |
| "eval_loss": 10.258369445800781, | |
| "eval_runtime": 420.9662, | |
| "eval_samples_per_second": 0.582, | |
| "eval_steps_per_second": 0.582, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.5872032193158954, | |
| "grad_norm": 1.32246732711792, | |
| "learning_rate": 3.772572564296005e-05, | |
| "loss": 10.3805, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 0.5975050301810865, | |
| "grad_norm": 0.46579450368881226, | |
| "learning_rate": 3.612985456190778e-05, | |
| "loss": 10.7162, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.5975050301810865, | |
| "eval_loss": 10.689937591552734, | |
| "eval_runtime": 393.2211, | |
| "eval_samples_per_second": 0.623, | |
| "eval_steps_per_second": 0.623, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.6078068410462777, | |
| "grad_norm": 0.3314819037914276, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 10.7026, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 0.6181086519114688, | |
| "grad_norm": 0.7479149699211121, | |
| "learning_rate": 3.298534127791785e-05, | |
| "loss": 10.4785, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.6181086519114688, | |
| "eval_loss": 10.441695213317871, | |
| "eval_runtime": 587.9582, | |
| "eval_samples_per_second": 0.417, | |
| "eval_steps_per_second": 0.417, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.6284104627766599, | |
| "grad_norm": 0.45178794860839844, | |
| "learning_rate": 3.144013755408895e-05, | |
| "loss": 10.4069, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 0.6387122736418511, | |
| "grad_norm": 0.5428746938705444, | |
| "learning_rate": 2.991522876735154e-05, | |
| "loss": 10.023, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6387122736418511, | |
| "eval_loss": 9.624427795410156, | |
| "eval_runtime": 394.3213, | |
| "eval_samples_per_second": 0.621, | |
| "eval_steps_per_second": 0.621, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6490140845070422, | |
| "grad_norm": 1.6642308235168457, | |
| "learning_rate": 2.8412282383075363e-05, | |
| "loss": 9.6059, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 0.6593158953722335, | |
| "grad_norm": 0.7868029475212097, | |
| "learning_rate": 2.693294185106562e-05, | |
| "loss": 9.2272, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.6593158953722335, | |
| "eval_loss": 8.930831909179688, | |
| "eval_runtime": 559.2753, | |
| "eval_samples_per_second": 0.438, | |
| "eval_steps_per_second": 0.438, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.6696177062374246, | |
| "grad_norm": 0.8221493363380432, | |
| "learning_rate": 2.547882480847461e-05, | |
| "loss": 9.15, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.6799195171026157, | |
| "grad_norm": 0.8017754554748535, | |
| "learning_rate": 2.405152131093926e-05, | |
| "loss": 9.1518, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6799195171026157, | |
| "eval_loss": 9.226869583129883, | |
| "eval_runtime": 395.8334, | |
| "eval_samples_per_second": 0.619, | |
| "eval_steps_per_second": 0.619, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6902213279678069, | |
| "grad_norm": 0.8554782867431641, | |
| "learning_rate": 2.2652592093878666e-05, | |
| "loss": 9.1271, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 0.700523138832998, | |
| "grad_norm": 0.8649684190750122, | |
| "learning_rate": 2.128356686585282e-05, | |
| "loss": 9.1733, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.700523138832998, | |
| "eval_loss": 9.243361473083496, | |
| "eval_runtime": 580.204, | |
| "eval_samples_per_second": 0.422, | |
| "eval_steps_per_second": 0.422, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.7108249496981891, | |
| "grad_norm": 0.5342640280723572, | |
| "learning_rate": 1.9945942635848748e-05, | |
| "loss": 9.5578, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 0.7211267605633803, | |
| "grad_norm": 0.5874167680740356, | |
| "learning_rate": 1.8641182076323148e-05, | |
| "loss": 9.3347, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7211267605633803, | |
| "eval_loss": 9.283143043518066, | |
| "eval_runtime": 401.3056, | |
| "eval_samples_per_second": 0.611, | |
| "eval_steps_per_second": 0.611, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7314285714285714, | |
| "grad_norm": 0.9126715064048767, | |
| "learning_rate": 1.7370711923791567e-05, | |
| "loss": 9.4285, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 0.7417303822937625, | |
| "grad_norm": 1.4541490077972412, | |
| "learning_rate": 1.6135921418712956e-05, | |
| "loss": 9.468, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7417303822937625, | |
| "eval_loss": 9.104644775390625, | |
| "eval_runtime": 406.3059, | |
| "eval_samples_per_second": 0.603, | |
| "eval_steps_per_second": 0.603, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7520321931589538, | |
| "grad_norm": 1.042210578918457, | |
| "learning_rate": 1.4938160786375572e-05, | |
| "loss": 8.9945, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 0.7623340040241449, | |
| "grad_norm": 1.5759915113449097, | |
| "learning_rate": 1.3778739760445552e-05, | |
| "loss": 8.9402, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.7623340040241449, | |
| "eval_loss": 9.010215759277344, | |
| "eval_runtime": 586.8102, | |
| "eval_samples_per_second": 0.418, | |
| "eval_steps_per_second": 0.418, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.772635814889336, | |
| "grad_norm": 1.4896786212921143, | |
| "learning_rate": 1.2658926150792322e-05, | |
| "loss": 9.0846, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7829376257545272, | |
| "grad_norm": 1.1868205070495605, | |
| "learning_rate": 1.157994445715706e-05, | |
| "loss": 9.1051, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.7829376257545272, | |
| "eval_loss": 9.261675834655762, | |
| "eval_runtime": 416.6181, | |
| "eval_samples_per_second": 0.588, | |
| "eval_steps_per_second": 0.588, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.7932394366197183, | |
| "grad_norm": 1.3363028764724731, | |
| "learning_rate": 1.0542974530180327e-05, | |
| "loss": 9.2574, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 0.8035412474849094, | |
| "grad_norm": 1.2219579219818115, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 9.2223, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.8035412474849094, | |
| "eval_loss": 9.392084121704102, | |
| "eval_runtime": 602.8211, | |
| "eval_samples_per_second": 0.406, | |
| "eval_steps_per_second": 0.406, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.8138430583501006, | |
| "grad_norm": 0.9515023827552795, | |
| "learning_rate": 8.599558442598998e-06, | |
| "loss": 9.4958, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 0.8241448692152917, | |
| "grad_norm": 0.989476203918457, | |
| "learning_rate": 7.695237378953223e-06, | |
| "loss": 9.3359, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.8241448692152917, | |
| "eval_loss": 9.327733993530273, | |
| "eval_runtime": 392.6991, | |
| "eval_samples_per_second": 0.624, | |
| "eval_steps_per_second": 0.624, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.834446680080483, | |
| "grad_norm": 1.3227804899215698, | |
| "learning_rate": 6.837175952121306e-06, | |
| "loss": 9.4079, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 0.8447484909456741, | |
| "grad_norm": 0.998881995677948, | |
| "learning_rate": 6.026312439675552e-06, | |
| "loss": 9.1508, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.8447484909456741, | |
| "eval_loss": 9.275510787963867, | |
| "eval_runtime": 607.3633, | |
| "eval_samples_per_second": 0.403, | |
| "eval_steps_per_second": 0.403, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.8550503018108652, | |
| "grad_norm": 0.7702223062515259, | |
| "learning_rate": 5.263533508961827e-06, | |
| "loss": 9.2486, | |
| "step": 332 | |
| }, | |
| { | |
| "epoch": 0.8653521126760564, | |
| "grad_norm": 0.594261646270752, | |
| "learning_rate": 4.549673247541875e-06, | |
| "loss": 9.5364, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8653521126760564, | |
| "eval_loss": 9.303129196166992, | |
| "eval_runtime": 403.5636, | |
| "eval_samples_per_second": 0.607, | |
| "eval_steps_per_second": 0.607, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8756539235412475, | |
| "grad_norm": 1.039624810218811, | |
| "learning_rate": 3.885512251130763e-06, | |
| "loss": 9.3301, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.8859557344064386, | |
| "grad_norm": 0.9027475118637085, | |
| "learning_rate": 3.271776770026963e-06, | |
| "loss": 9.4429, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.8859557344064386, | |
| "eval_loss": 9.322930335998535, | |
| "eval_runtime": 357.9628, | |
| "eval_samples_per_second": 0.684, | |
| "eval_steps_per_second": 0.684, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.8962575452716298, | |
| "grad_norm": 0.9633539319038391, | |
| "learning_rate": 2.7091379149682685e-06, | |
| "loss": 9.3991, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 0.9065593561368209, | |
| "grad_norm": 1.729811191558838, | |
| "learning_rate": 2.1982109232821178e-06, | |
| "loss": 9.3958, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.9065593561368209, | |
| "eval_loss": 9.340832710266113, | |
| "eval_runtime": 172.4755, | |
| "eval_samples_per_second": 1.42, | |
| "eval_steps_per_second": 1.42, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.916861167002012, | |
| "grad_norm": 0.8728967905044556, | |
| "learning_rate": 1.7395544861325718e-06, | |
| "loss": 9.4181, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 0.9271629778672033, | |
| "grad_norm": 0.8314731121063232, | |
| "learning_rate": 1.333670137599713e-06, | |
| "loss": 9.3778, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.9271629778672033, | |
| "eval_loss": 9.357659339904785, | |
| "eval_runtime": 122.9327, | |
| "eval_samples_per_second": 1.993, | |
| "eval_steps_per_second": 1.993, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.9374647887323944, | |
| "grad_norm": 1.05840003490448, | |
| "learning_rate": 9.810017062595322e-07, | |
| "loss": 9.5264, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 0.9477665995975855, | |
| "grad_norm": 1.1095401048660278, | |
| "learning_rate": 6.819348298638839e-07, | |
| "loss": 9.1859, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9477665995975855, | |
| "eval_loss": 9.360688209533691, | |
| "eval_runtime": 162.1198, | |
| "eval_samples_per_second": 1.511, | |
| "eval_steps_per_second": 1.511, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9580684104627767, | |
| "grad_norm": 0.8359835743904114, | |
| "learning_rate": 4.367965336512403e-07, | |
| "loss": 9.2367, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 0.9683702213279678, | |
| "grad_norm": 1.236433506011963, | |
| "learning_rate": 2.458548727494292e-07, | |
| "loss": 9.4256, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.9683702213279678, | |
| "eval_loss": 9.362174987792969, | |
| "eval_runtime": 122.9652, | |
| "eval_samples_per_second": 1.992, | |
| "eval_steps_per_second": 1.992, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.9786720321931589, | |
| "grad_norm": 0.7953875660896301, | |
| "learning_rate": 1.0931863906127327e-07, | |
| "loss": 9.2804, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.9889738430583501, | |
| "grad_norm": 1.0977773666381836, | |
| "learning_rate": 2.7337132953697554e-08, | |
| "loss": 9.3454, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 0.9889738430583501, | |
| "eval_loss": 9.3620023727417, | |
| "eval_runtime": 167.8824, | |
| "eval_samples_per_second": 1.459, | |
| "eval_steps_per_second": 1.459, | |
| "step": 384 | |
| } | |
| ], | |
| "logging_steps": 4, | |
| "max_steps": 388, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 8, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.5203300693467136e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |